subscore: Computing Subscores in Classical Test Theory and Item Response Theory

Functions for computing test subscores using different methods in both classical test theory (CTT) and item response theory (IRT). This package enables three types of subscoring methods within the framework of CTT and IRT, including (1) Wainer's augmentation method (Wainer et. al., 2001) <doi:10.4324/9781410604729>, (2) Haberman's subscoring methods (Haberman, 2008) <doi:10.3102/1076998607302636>, and (3) Yen's objective performance index (OPI; Yen, 1987) <https://www.ets.org/research/policy_research_reports/publications/paper/1987/hrap>. It also includes functions to compute Proportional Reduction of Mean Squared Errors (PRMSEs) in Haberman's methods which are used to examine whether test subscores are of added value. In addition, the package includes a function to assess the local independence assumption of IRT with Yen's Q3 statistic (Yen, 1984 <doi:10.1177/014662168400800201>; Yen, 1993 <doi:10.1111/j.1745-3984.1993.tb00423.x>).

install.packages("subscore")

Try the subscore package in your browser